Cloudbric Sets High Standards With Deep Learning Technology VISION
(PRWEB) February 19, 2019 -- Cloudbric, a provider of cloud-based web security services, is pleased to announce it had filed a patent application with the US Patent and Trademark Office (USPTO) for an innovative deep learning technology known as VISION for the cybersecurity market.
With the growing number of web applications users and cloud services, application layer cyberattacks are becoming one of the biggest challenges for cybersecurity. Fortunately, these challenges are well suited for the application of machine learning, a component of artificial intelligence. Cybersecurity vendors typically rely on rule-based techniques to identify malicious behavior, but this often leads to a high rate of false positives and thus low accuracy. On the other hand, machine learning makes it possible to identify anomalies and can adapt to changing behavior in real time.
Cloudbric aims to utilize deep learning, a subset of machine learning, to detect cyberattacks with outstanding accuracy. Combining with Cloudbric’s existing core technology, which is able to detect unknown and modified attacks and boasts one of the lowest false positive rates in the market, will make for a threat detection engine with even greater accuracy capabilities. The soon to be patented technology is based on an approach previously unexplored within the machine learning space and which will prove crucial for the cybersecurity industry that is besieged with ever-changing cyber threats that are becoming more frequent, sophisticated, and persistent.
Cloudbric’s deep learning machine will effectively learn and analyze hexadecimal characters to cover all URL patterns to determine whether traffic is malicious or not. Its fundamental advantage is its ability to overcome limitations of deep learning machines that are constrained to only learning by images or 68 characters — which is not enough to cover complex cyber threat patterns that may contain special characters. Early testing has shown an 85% accuracy rate increase compared to Cloudbric’s existing threat detection engine, and the team is experimenting with incremental learning in order to improve machine accuracy. For more, please refer to the Deep Learning Technical Overview at http://www.cloudbric.io.
VISION will be integrated into their web application firewall detection system to increase the accuracy of cyber threat identification. It will work in conjunction with their current filtering system to block incoming threats. Furthermore, this deep learning technology will be intreated into Cloudbric’s in-development mobile device security application to provide accurate, real-time protection against malware, spam, viruses, and phishing URLs found across the web.
“Since the launch of our blockchain-based cybersecurity project, we have continued to innovate and change the ways people look at cybersecurity and blockchain. We are pleased to strengthen our intellectual property protection with a patent covering an additional application of our core technology,” stated TJ Jung, Chief Executive Officer of Cloudbric. “We’re elated to be part of an exciting time where cybersecurity, blockchain, and AI can converge and integrate.”
Cloudbric officially submitted a patent for their deep learning module last year and is currently awaiting approval. Cloudbric also submitted a patent application at the Japan Patent Office for wider global recognition.
About Cloudbric
Cloudbric is a decentralized universal security platform featuring enterprise-grade website, mobile device and crypto asset security, as well as access to a community driven threat intelligence database. Visit http://www.cloudbric.com to begin using their website services today.
Karen Cruz, Cloudbric Corporation, https://www.cloudbric.com/, +82 221256735, [email protected]
Share this article